In the model in this paper, this happens for some (but not necessarily all) job switching agents. Therefore, the government faces a trade-off between the size of the increase in total production efficiency and the size of overcompensation in the transfer program.
Equivalence of National Income with Gross National Product
The proportional relationship in (15) arises because the production functions for the two sectors are Cobb-Douglas and symmetric.
Goods-market Equilibrium
After the partition of the type space given by (5), let ΘX(p0) and ΘY(p0) represent the prior partition of the type space into a subspace for X producers and another for Y producers, and let ΘX(p1) and ΘY (p1) represent each ex subpartition. Proposition 1 The economic welfare of the jobless agents will improve (respectively worsen) in accordance with the increase (respectively decrease) in the price of the goods they produce. The value of the profit function for agent(θ,τ) when the agent works for sectorY can be written as.
The fortunes and misfortunes of job changers change along with the slope of the beam from the origin. If the slope to the agent is steeper than that of the zero profit line and the agent is below the ex post division of labor line, τ = (p1)1−a2 ·θ, then the welfare of the agent is worsened. This σ2-equilibrium can also be called a σ-equilibrium, since the result of the second phase is also the final result of the entire compensation scheme.
The purpose of this hypothetical construction of the first stage is to see if the scheme can ensure Pareto gains (ex post) by moving as close as possible to an equilibrium where all the individual actors in the economy are as well off as they was before. Note that our definitions of overcompensation and rent neutrality represent two sides of the same coin. Despite the fact that much of the literature (on mechanism design) discusses the concept of "feasibility" in terms of non-negativity of state budgets (self-financing property), this paper separates the state budget issues (discussed above) from the information problems.
In this article, a plan is feasible when government policy instruments are based on observable (or at least taxable) variables. Now that we have defined all the necessary properties of the compensation scheme and examined the relevant results, we examine the results of possible compensation schemes.
Unanticipated Schemes
In terms of figure 2, this means, for example, that the government cannot distinguish between pointsq and nr, because in equilibrium the individuals at these points earn the same profit and produce the same amount of productX. The left side of the figure contains lines that represent the same percentage change in the profit or loss from trading. The right side contains lines indicating that those individuals make the same amount of residual profit.
This is because people who change jobs look the same if they earn the same profit, and are therefore represented by the same iso-current profit line. Of those who gain the same gain, it is the individuals at the top of the iso-current profit line who benefit the least (lose the most) from the change. Because the policymaker cannot distinguish between individuals on the same profit line, the policymaker must compensate all individuals on the same profit line at the same level as the least fortunate of those individuals, who are at the top of that line.
However, apart from the least advantaged individual, individuals who receive the same amount of compensation from the policy maker will get positive rents because their iso-percentage gain/loss lines are higher than those of the individual at the top end. Let us examine two points q and rin Figure 2, which are on the same iso-current profit line. Even though the individual is a winner, he or she receives the same amount of subsidy (as opposed to paying a tax) as the individual at point q.
Anticipated Schemes
In the first-stage equilibrium, the policymaker tries to make agents at least as good off as they were ex ante.20 Any non-negative revenue that accrues to the government can be returned to the agents in the second stage. There will be practically no job changers in this regard due to the change in trade conditions. This offer by the government guarantees that no one will be made worse off by trade liberalization, which holds because ex ante producers of Y have the opportunity to stay in the same industry and earn the same return as before.
Therefore, despite the subsidy for remaining employed in sector Y, the agents find it profitable to change occupations, provided the tax exemption is in the new sector. With regard to the job changers, who have created an overcompensation problem in the unexpected event, this scheme either taxes some or exempts some from tax;. In the extreme case, the policy maker must offer exactly the same tax subsidy rates as applied in the unexpected post-liberalization compensation scheme if the government maximizes the number of job changers.
The preceding analysis has shown that, in the case of a projected compensation scheme in which the government aims to achieve a Pareto improvement after the change, there is a trade-off between the aggregate output gains from trade and the amount of overcompensation. Thus, the survey subsidy for individuals switching jobs (as proposed by Feenstra and Lewis) may not be desirable in the context of our model. This paper has provided a model of individuals' occupational choices and welfare changes when the economy faces a change in the terms of trade, particularly in the case of trade liberalization.
Although the analysis in this paper can explain the long-term gains and losses of individuals from moving to a new sector, the model does not take into account the short-term costs of labor adjustment. Therefore, the main theoretical result of the paper – that no positive subsidy should be given to individuals who change jobs under a self-financing compensation scheme – should not be taken too literally. The compensation provided by the United States Department of Labor through its TAA program actually includes a relocation subsidy for those who move to a new location when they change jobs in response to trade changes.
Such a program can be justified to the extent that there are short-term frictional costs associated with job switching. A simplifying assumption made in this paper is that occupational talents are exogenously given to each individual. We have omitted the possibility of such dynamic development of individual talents through human capital investment.
Grossman and Shapiro (1982) analyzed the determinants of individual talent training when individual agents are identical ex ante. An interesting extension of this paper's model would be to include the dynamic formation of specific factors allowing investment in individual professional talents. This is a promising avenue for future research, and Ichida (2011) is one of the first forays down the road.
Proof of Lemma 1
Proof of Proposition 1
Vτ0(p)<0 can be proved similarly. 40) Because both Ka(1−a) and θ are nonnegative, the sign of the derivative of p.
Analysis of Profit Taxation System
Strict concavity of the production function X(·,·) guarantees that the second-order condition for the regular problem (42) holds, with strict inequality. So unless the profit tax rate falls by more than 1% when profits simultaneously rise by 1%, the agent maximizes profits even after profits have been taxed.
Proof of Proposition 4
Let τ(θ∗) be the lower bound for the value of the component τ in a setICP(θ∗), and let τ(θ∗) be the upper bound for the same subspace. Because all individuals in the set TKP(θ∗) are job changers from sector Y to sector X, they are currently producing output X. Since all members of the set TKP(θ∗) have the same talent,θ∗, their profit is the same: πX( p, r(p),θ∗).
Whether the individual who has the talentθ∗ wins or loses, and what the gain or loss is, depends on the value ofτ. Among those belonging to setICP(θ∗), there are many individuals who have the latent talent τ in the interval h. A politician who wants to ensure that Pareto wins from the economic change must be sure to do at least mediated individual as well as he or she was under the preceding situation.
Note also that this least wealthy individual must have had the most talent in the previous sector, Y, and therefore must have been the one with the most latent talent, τ(θ∗). With the exception of the individual at the point (θ∗,τ(θ∗)), which acts as zero, all individuals in the set ICP(θ∗) are overcompensated, since the inequality. The difference between the right and left sides of inequality (54) refers to the total amount of excess compensation for individuals who change jobs.
Unit-square Case: Unanticipated
The difference between the right and left sides of the inequality (54) is related to the total amount of overcompensation for job-changing individuals.. the parameter p) to be equal to real income, it is easy to find the tax -subsidy rates, for all groups, against which everyone is as well off as they were beforehand. Note that the tax subsidy rates must be based on observable variables (or variables that can be easily calculated). The characteristics of the tax subsidy rates for each group are therefore the following:. i) (linear) factor (commodity) tax on generic factors;. ii) (linear) profit tax on the occupational rewards for working producers of output X;. iii) (non-linear) profit tax on the occupational rewards for job-changing producers of output X;. iv) (non-linear) profit subsidy on the occupational rewards for job exchange producers of outputX; and. v) (linear) profit subsidy on the occupational rewards for continuing producers of outputY.
The linear factor tax for owners of generic factors is the same as in the best case. Given the above categorization, we create a further partition of the ability vector space as follows. The groups of permanent employees, ΘXX and ΘY Y, face the same linear tax subsidy scheme as in the best case.
So we focus on the lane changers, H, M, and L, all of which currently produce outputX. This completes the description of the tax subsidy scheme for the first-stage equilibrium in the unit square case. 2001): “Quasi-specific factors: Comparative advantage of workers in the two-sector production model”, Journal of International Economics.
H = all gainers (tax)
H = all winners (positive tax)
L = the group of individuals who were winning job-switchers ex ante
